1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144 | import sys
sys.stderr = open(snakemake.log[0], "w")
from sklearn import linear_model
from typing import List, Tuple
from scipy import stats
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
plt.style.use("ggplot")
DELIM = ","
PAIR_IDX = ("sample1", "sample2")
BLACK = "#4c566a"
BLUE = "#5e81ac"
XCOL = "COMPASS_dist"
YCOL = "ont_dist"
def robust_regression(x: List[float], y: List[float]) -> List[float]:
"""Returns the prediction of the model. This prediction is used with X to plot
line of best fit and get equation for that line.
"""
X = [[v] for v in x]
ransac = linear_model.RANSACRegressor()
ransac.fit(X, y)
pred = ransac.predict(X)
return pred
def fit_model(
x: List[float], y_pred: List[float]
) -> Tuple[float, float, float, float, float]:
"""Returns: gradient, intercept, r_value, p_value, std_err"""
return stats.linregress(x, y_pred)
class AsymmetrixMatrixError(Exception):
pass
def load_matrix(fpath, delim: str = DELIM, name: str = "") -> pd.DataFrame:
matrix = []
with open(fpath) as instream:
header = next(instream).rstrip()
names = np.array(header.split(delim)[1:])
idx = np.argsort(names)
sorted_names = names[idx]
for row in map(str.rstrip, instream):
# sort row according to the name sorting
sorted_row = np.array(row.split(delim)[1:], dtype=int)[idx]
matrix.append(sorted_row)
sorted_matrix = np.array(matrix)[idx]
n_samples = len(sorted_names)
diagonal_is_zero = all(sorted_matrix[i, i] == 0 for i in range(n_samples))
if not diagonal_is_zero:
raise AsymmetrixMatrixError("Distance matrix diagonal is not all zero")
matrix_is_symmetric = np.allclose(sorted_matrix, sorted_matrix.T)
if not matrix_is_symmetric:
raise AsymmetrixMatrixError("Distance matrix is not symmetric")
mx = pd.DataFrame(sorted_matrix, columns=sorted_names, index=sorted_names)
# remove the lower triangle of the matrix and the middle diagonal
mx = mx.where(np.triu(np.ones(mx.shape), k=1).astype(bool))
mx = mx.stack().rename(name).astype(int)
mx = mx.rename_axis(PAIR_IDX)
return mx
# load the data
compass_df = load_matrix(snakemake.input.compass_matrix, name=XCOL)
ont_df = load_matrix(snakemake.input.ont_matrix, name=YCOL)
# merge the matrices
data = pd.concat([compass_df, ont_df], axis=1)
data = data.reset_index().rename(
columns={"level_0": PAIR_IDX[0], "level_1": PAIR_IDX[1]}
)
fig, ax = plt.subplots(figsize=snakemake.params.figsize, dpi=snakemake.params.dpi)
# plot the full set of pairs
kwargs = snakemake.params.scatter_kws
kwargs["alpha"] /= 2
sns.scatterplot(data=data, x=XCOL, y=YCOL, ax=ax, **kwargs)
# fit robust linear regression
xs = data[XCOL]
ys = data[YCOL]
y_pred = robust_regression(xs, ys)
slope, intercept, *_ = fit_model(xs, y_pred)
equation = f"y={slope:.3f}x+{intercept:.3f}"
# plot line of best fit as defined by the model
sns.lineplot(x=xs, y=y_pred, ax=ax, label=equation, color=BLUE)
# plot identity line
sns.lineplot(x=xs, y=xs, ax=ax, label="y=x", color=BLACK, linestyle="--")
ax.set_xlabel(snakemake.params.xlabel)
ax.set_ylabel(snakemake.params.ylabel)
ax.legend(loc=snakemake.params.legend_loc)
# set the lower left corner of the zoom inset (x0, y0, width, height)
axins = ax.inset_axes(bounds=snakemake.params.inset_bounds)
# sub region of the original image
threshold = snakemake.params.inset_threshold
inset_data = data.query(f"{XCOL} <= @threshold")
kwargs["alpha"] *= 2
axins = sns.scatterplot(data=inset_data, x=XCOL, y=YCOL, ax=axins, **kwargs)
# plot line of best fit as defined by the model
xs = inset_data[XCOL]
ys = inset_data[YCOL]
y_pred = robust_regression(xs, ys)
slope, intercept, *_ = fit_model(xs, y_pred)
equation = f"y={slope:.3f}x+{intercept:.3f}"
# plot line of best fit as defined by the model
sns.lineplot(x=xs, y=y_pred, ax=axins, label=equation, color=BLUE)
# plot identity line
sns.lineplot(x=xs, y=xs, ax=axins, label="y=x", color=BLACK, linestyle="--")
# set the inset axis limits
axins.set_xlim(auto=True)
axins.set_ylim(auto=True)
# remove axis labels on inset
axins.set_xlabel("")
axins.set_ylabel("")
# styling of the inset zoom lines
ax.indicate_inset_zoom(axins, alpha=0.8, edgecolor=BLACK)
axins.tick_params(color=BLACK, labelcolor=BLACK)
# add a border to the inset
for spine in axins.spines.values():
spine.set_edgecolor(BLACK)
spine.set_alpha(0.5)
axins.legend(loc=snakemake.params.inset_legend_loc)
fig.savefig(snakemake.output.plot)
|